SkillFoxx

CLI FAQ Shortcuts skill

Skill that reads your prompt history, finds the asks you repeat and turns each into a short command such as /sent

  • 134 ★ on the repository
  • Last commit
  • Apache-2.0
Skill

Extended access

Why this level

  • Reads local Claude Code, Codex and Cursor prompt history, which can contain keys and passwords
  • Creates command files in the project at your choice
All reasons and checks

kishormorol/cli-faq-shortcutsAuthor:kishormorol

Install

In your terminal, with SkillFoxx CLI

npx skillfoxx add skills/cli-faq-shortcuts

Detects the agents on your machine, shows the access level and pins the version.

Other ways to install

Access and checks

How we review

Extended accessThe tool runs code or makes network calls and only reads external data.

  • Reads local Claude Code, Codex and Cursor prompt history, which can contain keys and passwords
  • Creates command files in the project at your choice

This is third-party code. Review the repository files before installing.

What it does

The `faq-shortcuts` skill reads your request history in Claude Code, Codex and Cursor, groups repeated asks and shows a table with counts. You pick rows and each becomes a short project command such as `/sent` or `/project billing`. On one real project the author found a single ask typed 230 times in different words. Extraction is done by `extract_asks.py`, standard-library Python with no packages, keys or network calls. Keys, tokens and passwords from old prompts are replaced with `[REDACTED]` before the agent sees the list. Grouping by intent is the agent's job. The `--since` and `--source` flags narrow the period and the source.

Who it is for. People who type the same long requests to Claude Code, Codex or Cursor every day and want short commands instead.

Good fit when

  • A project has built up history and you want to see the most frequent asks with counts
  • You want slash commands for one specific project, such as `/project billing`
  • Before publishing a skill or note you want to confirm secrets from old prompts are hidden
  • You want only a recent period or only requests made to Codex

Not a fit when

  • The prompt history is empty or the project is new: there are no repeated asks yet
  • You need requests from browser chats: the skill reads only local Claude Code, Codex and Cursor history
  • You work on Windows and do not want to set up the install by hand: the main path targets macOS and Linux

Example request

What do I keep asking in this project? Show a table of the most frequent asks and propose short commands

Limitations

  • The main install is described for macOS and Linux and needs `git` and Python 3.9 or newer. A separate README section covers Windows.
  • The repository is called `cli-faq-shortcuts` while the installed skill is `faq-shortcuts`.
  • The exact-repeat counting script undercounts real repetition: grouping by intent is the agent's job, and quality depends on the model.
  • Secrets are hidden by a heuristic at extraction, and the final commands remain your choice.
  • The README demo video uses synthetic data and is not a live run.

How to disable. Delete `~/.agents/skills/faq-shortcuts` and the `~/.claude/skills/faq-shortcuts` link. Commands created in a project live in the project folder and are removed by hand.

FAQ

How do I run the skill?

Open the project in Claude Code, Codex or Cursor in a new session and ask what you keep asking, or type `/faq-shortcuts` in Claude Code.

Do my prompts leave my machine?

The extraction script makes no network calls and uses only the Python standard library. The list is then read by your agent like any other session text.

What does the --source flag do?

It limits the sample to one history source, for example `--source codex`. `--since 2026-06-01` drops older requests.

How does the skill hide secrets?

Keys, tokens and passwords in old prompts are replaced with `[REDACTED]` before the list reaches the agent. The method is heuristic, so skim the final list.

How can I see repeat counts without the agent?

Run `python3 ~/.agents/skills/faq-shortcuts/scripts/extract_asks.py .` and pipe it through `sort | uniq -c`. This counts exact repeats only, so the number is an undercount.

Discussion

No questions or reviews yet.

Skills

A skill that builds an anonymized developer profile and a vertical poster from local Claude Code and other CLI history

SkillExtended access175

claude-history searches and opens past Claude Code, Codex and Pi conversations in the terminal and gives Claude a history skill

CLIExtended access542

/bro

bro

The /bro command re-explains the assistant's last reply in plain language while commands and numbers stay as they were.

SkillMinimal access452

A CLI that gathers Claude Code and Codex session history into a local graph, searches it and turns repeated mistakes into fixes

CLIExtended access123

About a hundred skills for Claude Code: meetings, research, publishing, TDD, Telegram, email and personal analytics

PluginBroad access413

A skill that analyzes materials and subtitles, checks facts against sources and keeps a personal library in a local folder.

SkillMinimal access98

README in short

The README opens with three benefit points and a short video. A three-step install, flag commands, a sample table and a Windows section follow.

README badge

Report a problem